[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: New Version Notification for draft-ietf-v6ops-v6inixp-02
* On Thu, 1 Oct 2009 13:53:23 +0200
* Martin Pels <firstname.lastname@example.org> wrote:
> On Tue, 8 Sep 2009 23:25:37 -0300
> Roque Gagliano <email@example.com> wrote:
> > Hi,
> > I issued a new ID of the draft with the changes that came up at the
> > Stockholm meeting. Changes were:
> > - I explained why ULA is not a good idea.
> > - I added that addressing can use two different /48, one for the
> > LANs and the second one for the internal services and not necessarily
> > one / 47, as comments at the meeting.
> In section 3 you added the following: "IPv6 prefixes for IXP LAN's are
> typically publicly well known.".
> I suggest changing this to something like: "IPv6 prefixes for IXP LANs
> are typically taken from dedicated IPv6 blocks for IXP assignments,
> reserved for this purpose by the different RIRs."
Thanks for your effort about IPv6 IXP operation.
If modifying as above, add notions that IXP can select PI address policy
either routed or un-routed in the global table.
- AS boundry monitoring IXP segment is possible.
- Attacking IXP segment from non IXP participants is possible.
- Attacking IXP segment from non IXP participants is impossible.
- AS boundry monitoring IXP segment is impossible.
Are there any other pros/cons?